qoherent/experiments-07-07-12-local-llm-knowledge-2
23
0
Fork 0
Code Issues Pull Requests Workflows Packages Projects Releases Wiki Activity
Push Tracker
experiments-07-07-12-local-.../results/gpt-oss-120b.server.log

930 lines
90 KiB
Plaintext
Raw Permalink Normal View History

G
readme report
2026-07-12 21:35:10 -04:00
0.00.065.006 I log_info: verbosity = 3 (adjust with the `-lv N` CLI arg)
0.00.065.009 I device_info:
0.00.065.027 I - ROCm0 : AMD Radeon Graphics (32752 MiB, 32728 MiB free)
0.00.065.031 I - ROCm1 : AMD Radeon Graphics (32752 MiB, 32728 MiB free)
0.00.065.037 I - CPU : AMD Ryzen 9 3900X 12-Core Processor (128730 MiB, 128730 MiB free)
0.00.065.096 I system_info: n_threads = 12 (n_threads_batch = 12) / 24 | ROCm : NO_VMM = 1 | PEER_MAX_BATCH_SIZE = 128 | CPU : SSE3 = 1 | SSSE3 = 1 | AVX = 1 | AVX2 = 1 | F16C = 1 | FMA = 1 | BMI2 = 1 | LLAMAFILE = 1 | OPENMP = 1 | REPACK = 1 |
0.00.065.099 I srv llama_server: n_parallel is set to auto, using n_parallel = 4 and kv_unified = true
0.00.065.136 I srv init: running without SSL
0.00.065.206 I srv init: using 23 threads for HTTP server
0.00.065.531 I srv start: binding port with default address family
0.00.066.730 I srv llama_server: loading model
0.00.066.739 I srv load_model: loading model '/home/qrf/.cache/llama.cpp/openai_gpt-oss-120b-Q4_K_M/openai_gpt-oss-120b-Q4_K_M-00001-of-00002.gguf'
0.00.066.786 I common_init_result: fitting params to device memory ...
0.00.066.790 I common_init_result: (for bugs during this step try to reproduce them with -fit off, or provide --verbose logs if the bug only occurs with -fit on)
0.01.515.237 W load: setting token '<|message|>' (200008) attribute to USER_DEFINED (16), old attributes: 8
0.01.515.244 W load: setting token '<|start|>' (200006) attribute to USER_DEFINED (16), old attributes: 8
0.01.515.245 W load: setting token '<|constrain|>' (200003) attribute to USER_DEFINED (16), old attributes: 8
0.01.516.306 W load: setting token '<|channel|>' (200005) attribute to USER_DEFINED (16), old attributes: 8
0.01.520.577 W load: special_eog_ids contains both '<|return|>' and '<|call|>', or '<|calls|>' and '<|flush|>' tokens, removing '<|end|>' token from EOG list
0.33.601.883 W llama_context: n_ctx_seq (8192) < n_ctx_train (131072) -- the full capacity of the model will not be utilized
0.33.834.736 I common_init_from_params: warming up the model with an empty run - please wait ... (--no-warmup to disable)
0.34.023.757 I srv load_model: initializing slots, n_slots = 4
0.34.070.409 W common_speculative_init: no implementations specified for speculative decoding
0.34.070.414 I slot load_model: id 0 | task -1 | new slot, n_ctx = 8192
0.34.070.417 I slot load_model: id 1 | task -1 | new slot, n_ctx = 8192
0.34.070.418 I slot load_model: id 2 | task -1 | new slot, n_ctx = 8192
0.34.070.418 I slot load_model: id 3 | task -1 | new slot, n_ctx = 8192
0.34.070.464 I srv load_model: prompt cache is enabled, size limit: 8192 MiB
0.34.070.465 I srv load_model: use `--cache-ram 0` to disable the prompt cache
0.34.070.465 I srv load_model: for more info see https://github.com/ggml-org/llama.cpp/pull/16391
0.34.070.465 I srv load_model: context checkpoints enabled, max = 32, min spacing = 256
0.34.070.484 I srv init: idle slots will be saved to prompt cache and cleared upon starting a new task
0.34.075.356 I init: chat template, example_format: '<|start|>system<|message|>You are ChatGPT, a large language model trained by OpenAI.
Knowledge cutoff: 2024-06
Current date: 2026-07-11
Reasoning: medium
# Valid channels: analysis, commentary, final. Channel must be included for every message.<|end|><|start|>developer<|message|># Instructions
You are a helpful assistant<|end|><|start|>user<|message|>Hello<|end|><|start|>assistant<|channel|>final<|message|>Hi there<|end|><|start|>user<|message|>How are you?<|end|><|start|>assistant'
0.34.075.917 I srv init: init: chat template, thinking = 1
0.34.075.955 I srv llama_server: model loaded
0.34.075.959 I srv llama_server: server is listening on http://127.0.0.1:8090
0.34.075.963 I srv update_slots: all slots are idle
0.35.001.855 I srv operator(): Chat format: peg-native
0.35.005.643 I slot get_availabl: id 3 | task -1 | selected slot by LRU, t_last = -1
0.35.005.647 I srv get_availabl: updating prompt cache
0.35.005.653 I srv load: - looking for better prompt, base f_keep = -1.000, sim = 0.000
0.35.005.658 I srv update: - cache state: 0 prompts, 0.000 MiB (limits: 8192.000 MiB, 8192 tokens, 8589934592 est)
0.35.005.659 I srv get_availabl: prompt cache update took 0.01 ms
0.35.006.597 I slot launch_slot_: id 3 | task 0 | processing task, is_child = 0
0.35.006.600 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
0.35.006.601 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
0.35.006.611 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
0.35.006.611 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
0.35.006.616 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
0.35.006.616 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
0.36.232.065 I slot create_check: id 3 | task 0 | created context checkpoint 1 of 32 (pos_min = 0, pos_max = 108, n_tokens = 109, size = 3.834 MiB)
0.38.874.562 I slot print_timing: id 3 | task 0 | prompt eval time = 1479.70 ms / 133 tokens ( 11.13 ms per token, 89.88 tokens per second)
0.38.874.565 I slot print_timing: id 3 | task 0 | eval time = 2388.22 ms / 150 tokens ( 15.92 ms per token, 62.81 tokens per second)
0.38.874.566 I slot print_timing: id 3 | task 0 | total time = 3867.92 ms / 283 tokens
0.38.874.571 I slot print_timing: id 3 | task 0 | graphs reused = 148
0.38.874.626 I slot release: id 3 | task 0 | stop processing: n_tokens = 282, truncated = 0
0.38.874.630 I srv update_slots: all slots are idle
0.38.883.423 I srv operator(): Chat format: peg-native
0.38.887.152 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.896 (> 0.100 thold), f_keep = 0.397
0.38.887.154 I srv get_availabl: updating prompt cache
0.38.887.189 W srv prompt_save: - saving prompt with length 282, total state size = 14.420 MiB (draft: 0.000 MiB)
0.38.895.511 I srv load: - looking for better prompt, base f_keep = 0.397, sim = 0.896
0.38.895.517 I srv update: - cache state: 1 prompts, 18.253 MiB (limits: 8192.000 MiB, 8192 tokens, 126560 est)
0.38.895.518 I srv update: - prompt 0x65411f701010: 282 tokens, checkpoints: 1, 18.253 MiB
0.38.895.519 I srv get_availabl: prompt cache update took 8.37 ms
0.38.896.445 I slot launch_slot_: id 3 | task 153 | processing task, is_child = 0
0.38.896.447 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
0.38.896.448 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
0.38.896.456 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
0.38.896.456 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
0.38.896.461 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
0.38.896.461 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
0.38.896.473 I slot update_slots: id 3 | task 153 | Checking checkpoint with [0, 108] against 0...
0.38.899.057 W slot update_slots: id 3 | task 153 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
0.40.270.462 I slot print_timing: id 3 | task 153 | prompt eval time = 181.70 ms / 17 tokens ( 10.69 ms per token, 93.56 tokens per second)
0.40.270.465 I slot print_timing: id 3 | task 153 | eval time = 1192.28 ms / 76 tokens ( 15.69 ms per token, 63.74 tokens per second)
0.40.270.465 I slot print_timing: id 3 | task 153 | total time = 1373.98 ms / 93 tokens
0.40.270.466 I slot print_timing: id 3 | task 153 | graphs reused = 222
0.40.270.498 I slot release: id 3 | task 153 | stop processing: n_tokens = 200, truncated = 0
0.40.270.505 I srv update_slots: all slots are idle
0.40.278.333 I srv operator(): Chat format: peg-native
0.40.282.038 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.882 (> 0.100 thold), f_keep = 0.560
0.40.283.067 I slot launch_slot_: id 3 | task 232 | processing task, is_child = 0
0.40.283.068 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
0.40.283.069 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
0.40.283.076 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
0.40.283.076 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
0.40.283.081 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
0.40.283.081 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
0.40.283.089 I slot update_slots: id 3 | task 232 | Checking checkpoint with [0, 108] against 0...
0.40.285.607 W slot update_slots: id 3 | task 232 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
0.42.304.520 I slot print_timing: id 3 | task 232 | prompt eval time = 198.22 ms / 19 tokens ( 10.43 ms per token, 95.85 tokens per second)
0.42.304.524 I slot print_timing: id 3 | task 232 | eval time = 1823.21 ms / 115 tokens ( 15.85 ms per token, 63.08 tokens per second)
0.42.304.525 I slot print_timing: id 3 | task 232 | total time = 2021.43 ms / 134 tokens
0.42.304.525 I slot print_timing: id 3 | task 232 | graphs reused = 335
0.42.304.558 I slot release: id 3 | task 232 | stop processing: n_tokens = 241, truncated = 0
0.42.304.567 I srv update_slots: all slots are idle
0.42.312.782 I srv operator(): Chat format: peg-native
0.42.316.602 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.904 (> 0.100 thold), f_keep = 0.469
0.42.316.604 I srv get_availabl: updating prompt cache
0.42.316.631 W srv prompt_save: - saving prompt with length 241, total state size = 12.978 MiB (draft: 0.000 MiB)
0.42.330.195 I srv load: - looking for better prompt, base f_keep = 0.469, sim = 0.904
0.42.330.203 I srv update: - cache state: 2 prompts, 35.065 MiB (limits: 8192.000 MiB, 8192 tokens, 122185 est)
0.42.330.204 I srv update: - prompt 0x65411f701010: 282 tokens, checkpoints: 1, 18.253 MiB
0.42.330.205 I srv update: - prompt 0x65411f8d6280: 241 tokens, checkpoints: 1, 16.811 MiB
0.42.330.206 I srv get_availabl: prompt cache update took 13.60 ms
0.42.331.152 I slot launch_slot_: id 3 | task 350 | processing task, is_child = 0
0.42.331.155 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
0.42.331.156 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
0.42.331.170 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
0.42.331.171 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
0.42.331.176 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
0.42.331.176 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
0.42.331.189 I slot update_slots: id 3 | task 350 | Checking checkpoint with [0, 108] against 0...
0.42.333.452 W slot update_slots: id 3 | task 350 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
0.44.567.460 I slot print_timing: id 3 | task 350 | prompt eval time = 184.30 ms / 17 tokens ( 10.84 ms per token, 92.24 tokens per second)
0.44.567.463 I slot print_timing: id 3 | task 350 | eval time = 2051.97 ms / 129 tokens ( 15.91 ms per token, 62.87 tokens per second)
0.44.567.464 I slot print_timing: id 3 | task 350 | total time = 2236.26 ms / 146 tokens
0.44.567.465 I slot print_timing: id 3 | task 350 | graphs reused = 462
0.44.567.497 I slot release: id 3 | task 350 | stop processing: n_tokens = 253, truncated = 0
0.44.567.505 I srv update_slots: all slots are idle
0.44.575.703 I srv operator(): Chat format: peg-native
0.44.579.402 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.883 (> 0.100 thold), f_keep = 0.447
0.44.579.405 I srv get_availabl: updating prompt cache
0.44.579.433 W srv prompt_save: - saving prompt with length 253, total state size = 13.400 MiB (draft: 0.000 MiB)
0.44.593.298 I srv load: - looking for better prompt, base f_keep = 0.447, sim = 0.883
0.44.593.303 I srv load: - found better prompt with f_keep = 0.477, sim = 0.898
0.44.599.118 I srv update: - cache state: 2 prompts, 35.487 MiB (limits: 8192.000 MiB, 8192 tokens, 123502 est)
0.44.599.121 I srv update: - prompt 0x65411f701010: 282 tokens, checkpoints: 1, 18.253 MiB
0.44.599.121 I srv update: - prompt 0x65411f8bbeb0: 253 tokens, checkpoints: 1, 17.233 MiB
0.44.599.122 I srv get_availabl: prompt cache update took 19.72 ms
0.44.600.045 I slot launch_slot_: id 3 | task 482 | processing task, is_child = 0
0.44.600.047 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
0.44.600.048 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
0.44.600.057 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
0.44.600.057 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
0.44.600.062 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
0.44.600.063 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
0.44.600.075 I slot update_slots: id 3 | task 482 | Checking checkpoint with [0, 108] against 0...
0.44.602.358 W slot update_slots: id 3 | task 482 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
0.45.920.542 I slot print_timing: id 3 | task 482 | prompt eval time = 192.00 ms / 20 tokens ( 9.60 ms per token, 104.17 tokens per second)
0.45.920.545 I slot print_timing: id 3 | task 482 | eval time = 1128.46 ms / 71 tokens ( 15.89 ms per token, 62.92 tokens per second)
0.45.920.546 I slot print_timing: id 3 | task 482 | total time = 1320.46 ms / 91 tokens
0.45.920.546 I slot print_timing: id 3 | task 482 | graphs reused = 531
0.45.920.578 I slot release: id 3 | task 482 | stop processing: n_tokens = 198, truncated = 0
0.45.920.586 I srv update_slots: all slots are idle
0.45.928.405 I srv operator(): Chat format: peg-native
0.45.932.089 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.875 (> 0.100 thold), f_keep = 0.566
0.45.933.086 I slot launch_slot_: id 3 | task 556 | processing task, is_child = 0
0.45.933.088 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
0.45.933.088 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
0.45.933.095 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
0.45.933.096 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
0.45.933.101 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
0.45.933.101 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
0.45.933.109 I slot update_slots: id 3 | task 556 | Checking checkpoint with [0, 108] against 0...
0.45.935.729 W slot update_slots: id 3 | task 556 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
0.47.562.530 I slot print_timing: id 3 | task 556 | prompt eval time = 199.44 ms / 20 tokens ( 9.97 ms per token, 100.28 tokens per second)
0.47.562.533 I slot print_timing: id 3 | task 556 | eval time = 1429.98 ms / 90 tokens ( 15.89 ms per token, 62.94 tokens per second)
0.47.562.534 I slot print_timing: id 3 | task 556 | total time = 1629.42 ms / 110 tokens
0.47.562.534 I slot print_timing: id 3 | task 556 | graphs reused = 619
0.47.562.565 I slot release: id 3 | task 556 | stop processing: n_tokens = 217, truncated = 0
0.47.562.573 I srv update_slots: all slots are idle
0.47.570.337 I srv operator(): Chat format: peg-native
0.47.574.025 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.848 (> 0.100 thold), f_keep = 0.516
0.47.574.996 I slot launch_slot_: id 3 | task 649 | processing task, is_child = 0
0.47.574.998 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
0.47.574.998 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
0.47.575.006 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
0.47.575.006 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
0.47.575.012 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
0.47.575.012 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
0.47.575.022 I slot update_slots: id 3 | task 649 | Checking checkpoint with [0, 108] against 0...
0.47.577.431 W slot update_slots: id 3 | task 649 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
0.49.367.571 I slot print_timing: id 3 | task 649 | prompt eval time = 248.56 ms / 24 tokens ( 10.36 ms per token, 96.56 tokens per second)
0.49.367.575 I slot print_timing: id 3 | task 649 | eval time = 1543.98 ms / 97 tokens ( 15.92 ms per token, 62.82 tokens per second)
0.49.367.576 I slot print_timing: id 3 | task 649 | total time = 1792.54 ms / 121 tokens
0.49.367.577 I slot print_timing: id 3 | task 649 | graphs reused = 714
0.49.367.612 I slot release: id 3 | task 649 | stop processing: n_tokens = 228, truncated = 0
0.49.367.620 I srv update_slots: all slots are idle
0.49.375.375 I srv operator(): Chat format: peg-native
0.49.379.056 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.926 (> 0.100 thold), f_keep = 0.496
0.49.379.058 I srv get_availabl: updating prompt cache
0.49.379.083 W srv prompt_save: - saving prompt with length 228, total state size = 12.521 MiB (draft: 0.000 MiB)
0.49.386.161 I srv load: - looking for better prompt, base f_keep = 0.496, sim = 0.926
0.49.386.173 I srv update: - cache state: 3 prompts, 51.841 MiB (limits: 8192.000 MiB, 8192 tokens, 120570 est)
0.49.386.174 I srv update: - prompt 0x65411f701010: 282 tokens, checkpoints: 1, 18.253 MiB
0.49.386.174 I srv update: - prompt 0x65411f8bbeb0: 253 tokens, checkpoints: 1, 17.233 MiB
0.49.386.175 I srv update: - prompt 0x65411f77c9c0: 228 tokens, checkpoints: 1, 16.354 MiB
0.49.386.176 I srv get_availabl: prompt cache update took 7.12 ms
0.49.387.109 I slot launch_slot_: id 3 | task 749 | processing task, is_child = 0
0.49.387.112 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
0.49.387.112 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
0.49.387.121 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
0.49.387.121 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
0.49.387.127 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
0.49.387.127 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
0.49.387.137 I slot update_slots: id 3 | task 749 | Checking checkpoint with [0, 108] against 0...
0.49.389.369 W slot update_slots: id 3 | task 749 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
0.51.033.580 I slot print_timing: id 3 | task 749 | prompt eval time = 168.46 ms / 14 tokens ( 12.03 ms per token, 83.10 tokens per second)
0.51.033.583 I slot print_timing: id 3 | task 749 | eval time = 1477.97 ms / 93 tokens ( 15.89 ms per token, 62.92 tokens per second)
0.51.033.583 I slot print_timing: id 3 | task 749 | total time = 1646.44 ms / 107 tokens
0.51.033.584 I slot print_timing: id 3 | task 749 | graphs reused = 805
0.51.033.617 I slot release: id 3 | task 749 | stop processing: n_tokens = 214, truncated = 0
0.51.033.625 I srv update_slots: all slots are idle
0.51.041.636 I srv operator(): Chat format: peg-native
0.51.045.314 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.848 (> 0.100 thold), f_keep = 0.523
0.51.046.295 I slot launch_slot_: id 3 | task 845 | processing task, is_child = 0
0.51.046.297 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
0.51.046.298 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
0.51.046.306 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
0.51.046.307 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
0.51.046.314 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
0.51.046.314 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
0.51.046.326 I slot update_slots: id 3 | task 845 | Checking checkpoint with [0, 108] against 0...
0.51.048.854 W slot update_slots: id 3 | task 845 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
0.52.602.585 I slot print_timing: id 3 | task 845 | prompt eval time = 250.71 ms / 24 tokens ( 10.45 ms per token, 95.73 tokens per second)
0.52.602.588 I slot print_timing: id 3 | task 845 | eval time = 1305.55 ms / 82 tokens ( 15.92 ms per token, 62.81 tokens per second)
0.52.602.589 I slot print_timing: id 3 | task 845 | total time = 1556.25 ms / 106 tokens
0.52.602.590 I slot print_timing: id 3 | task 845 | graphs reused = 885
0.52.602.624 I slot release: id 3 | task 845 | stop processing: n_tokens = 213, truncated = 0
0.52.602.632 I srv update_slots: all slots are idle
0.52.610.551 I srv operator(): Chat format: peg-native
0.52.614.203 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.882 (> 0.100 thold), f_keep = 0.526
0.52.615.106 I slot launch_slot_: id 3 | task 930 | processing task, is_child = 0
0.52.615.108 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
0.52.615.109 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
0.52.615.115 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
0.52.615.115 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
0.52.615.120 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
0.52.615.121 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
0.52.615.129 I slot update_slots: id 3 | task 930 | Checking checkpoint with [0, 108] against 0...
0.52.617.640 W slot update_slots: id 3 | task 930 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
0.55.351.764 I slot print_timing: id 3 | task 930 | prompt eval time = 196.80 ms / 19 tokens ( 10.36 ms per token, 96.55 tokens per second)
0.55.351.767 I slot print_timing: id 3 | task 930 | eval time = 2539.83 ms / 159 tokens ( 15.97 ms per token, 62.60 tokens per second)
0.55.351.768 I slot print_timing: id 3 | task 930 | total time = 2736.63 ms / 178 tokens
0.55.351.768 I slot print_timing: id 3 | task 930 | graphs reused = 1041
0.55.351.800 I slot release: id 3 | task 930 | stop processing: n_tokens = 285, truncated = 0
0.55.351.808 I srv update_slots: all slots are idle
0.55.360.088 I srv operator(): Chat format: peg-native
0.55.364.136 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.848 (> 0.100 thold), f_keep = 0.393
0.55.364.138 I srv get_availabl: updating prompt cache
0.55.364.178 W srv prompt_save: - saving prompt with length 285, total state size = 14.525 MiB (draft: 0.000 MiB)
0.55.378.790 I srv load: - looking for better prompt, base f_keep = 0.393, sim = 0.848
0.55.378.797 I srv update: - cache state: 4 prompts, 70.200 MiB (limits: 8192.000 MiB, 8192 tokens, 122296 est)
0.55.378.798 I srv update: - prompt 0x65411f701010: 282 tokens, checkpoints: 1, 18.253 MiB
0.55.378.799 I srv update: - prompt 0x65411f8bbeb0: 253 tokens, checkpoints: 1, 17.233 MiB
0.55.378.802 I srv update: - prompt 0x65411f77c9c0: 228 tokens, checkpoints: 1, 16.354 MiB
0.55.378.803 I srv update: - prompt 0x65411c32e000: 285 tokens, checkpoints: 1, 18.359 MiB
0.55.378.804 I srv get_availabl: prompt cache update took 14.66 ms
0.55.379.716 I slot launch_slot_: id 3 | task 1092 | processing task, is_child = 0
0.55.379.718 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
0.55.379.719 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
0.55.379.729 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
0.55.379.729 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
0.55.379.735 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
0.55.379.735 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
0.55.379.747 I slot update_slots: id 3 | task 1092 | Checking checkpoint with [0, 108] against 0...
0.55.382.207 W slot update_slots: id 3 | task 1092 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
0.57.450.623 I slot print_timing: id 3 | task 1092 | prompt eval time = 257.90 ms / 24 tokens ( 10.75 ms per token, 93.06 tokens per second)
0.57.450.626 I slot print_timing: id 3 | task 1092 | eval time = 1812.97 ms / 114 tokens ( 15.90 ms per token, 62.88 tokens per second)
0.57.450.627 I slot print_timing: id 3 | task 1092 | total time = 2070.87 ms / 138 tokens
0.57.450.628 I slot print_timing: id 3 | task 1092 | graphs reused = 1153
0.57.450.660 I slot release: id 3 | task 1092 | stop processing: n_tokens = 245, truncated = 0
0.57.450.668 I srv update_slots: all slots are idle
0.57.458.801 I srv operator(): Chat format: peg-native
0.57.462.552 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.875 (> 0.100 thold), f_keep = 0.457
0.57.462.554 I srv get_availabl: updating prompt cache
0.57.462.584 W srv prompt_save: - saving prompt with length 245, total state size = 13.118 MiB (draft: 0.000 MiB)
0.57.476.014 I srv load: - looking for better prompt, base f_keep = 0.457, sim = 0.875
0.57.476.019 I srv load: - found better prompt with f_keep = 0.496, sim = 0.883
0.57.481.667 I srv update: - cache state: 4 prompts, 70.798 MiB (limits: 8192.000 MiB, 8192 tokens, 123231 est)
0.57.481.670 I srv update: - prompt 0x65411f701010: 282 tokens, checkpoints: 1, 18.253 MiB
0.57.481.670 I srv update: - prompt 0x65411f8bbeb0: 253 tokens, checkpoints: 1, 17.233 MiB
0.57.481.671 I srv update: - prompt 0x65411c32e000: 285 tokens, checkpoints: 1, 18.359 MiB
0.57.481.671 I srv update: - prompt 0x65411f899dc0: 245 tokens, checkpoints: 1, 16.952 MiB
0.57.481.672 I srv get_availabl: prompt cache update took 19.12 ms
0.57.482.600 I slot launch_slot_: id 3 | task 1209 | processing task, is_child = 0
0.57.482.602 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
0.57.482.603 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
0.57.482.610 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
0.57.482.611 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
0.57.482.616 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
0.57.482.616 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
0.57.482.627 I slot update_slots: id 3 | task 1209 | Checking checkpoint with [0, 108] against 0...
0.57.484.848 W slot update_slots: id 3 | task 1209 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
0.58.992.643 I slot print_timing: id 3 | task 1209 | prompt eval time = 192.50 ms / 20 tokens ( 9.62 ms per token, 103.90 tokens per second)
0.58.992.647 I slot print_timing: id 3 | task 1209 | eval time = 1317.51 ms / 83 tokens ( 15.87 ms per token, 63.00 tokens per second)
0.58.992.648 I slot print_timing: id 3 | task 1209 | total time = 1510.01 ms / 103 tokens
0.58.992.649 I slot print_timing: id 3 | task 1209 | graphs reused = 1234
0.58.992.682 I slot release: id 3 | task 1209 | stop processing: n_tokens = 210, truncated = 0
0.58.992.691 I srv update_slots: all slots are idle
0.59.000.755 I srv operator(): Chat format: peg-native
0.59.004.624 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.868 (> 0.100 thold), f_keep = 0.533
0.59.005.559 I slot launch_slot_: id 3 | task 1295 | processing task, is_child = 0
0.59.005.561 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
0.59.005.561 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
0.59.005.569 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
0.59.005.570 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
0.59.005.575 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
0.59.005.575 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
0.59.005.583 I slot update_slots: id 3 | task 1295 | Checking checkpoint with [0, 108] against 0...
0.59.008.154 W slot update_slots: id 3 | task 1295 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.00.283.635 I slot print_timing: id 3 | task 1295 | prompt eval time = 200.62 ms / 21 tokens ( 9.55 ms per token, 104.67 tokens per second)
1.00.283.638 I slot print_timing: id 3 | task 1295 | eval time = 1077.42 ms / 68 tokens ( 15.84 ms per token, 63.11 tokens per second)
1.00.283.639 I slot print_timing: id 3 | task 1295 | total time = 1278.05 ms / 89 tokens
1.00.283.640 I slot print_timing: id 3 | task 1295 | graphs reused = 1300
1.00.283.674 I slot release: id 3 | task 1295 | stop processing: n_tokens = 196, truncated = 0
1.00.283.683 I srv update_slots: all slots are idle
1.00.291.762 I srv operator(): Chat format: peg-native
1.00.295.667 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.875 (> 0.100 thold), f_keep = 0.571
1.00.296.585 I slot launch_slot_: id 3 | task 1366 | processing task, is_child = 0
1.00.296.587 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.00.296.587 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.00.296.594 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.00.296.594 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.00.296.599 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.00.296.600 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.00.296.608 I slot update_slots: id 3 | task 1366 | Checking checkpoint with [0, 108] against 0...
1.00.299.149 W slot update_slots: id 3 | task 1366 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.01.728.686 I slot print_timing: id 3 | task 1366 | prompt eval time = 194.96 ms / 20 tokens ( 9.75 ms per token, 102.58 tokens per second)
1.01.728.690 I slot print_timing: id 3 | task 1366 | eval time = 1237.11 ms / 78 tokens ( 15.86 ms per token, 63.05 tokens per second)
1.01.728.691 I slot print_timing: id 3 | task 1366 | total time = 1432.07 ms / 98 tokens
1.01.728.692 I slot print_timing: id 3 | task 1366 | graphs reused = 1376
1.01.728.728 I slot release: id 3 | task 1366 | stop processing: n_tokens = 205, truncated = 0
1.01.728.738 I srv update_slots: all slots are idle
1.01.736.833 I srv operator(): Chat format: peg-native
1.01.740.548 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.862 (> 0.100 thold), f_keep = 0.546
1.01.741.399 I slot launch_slot_: id 3 | task 1447 | processing task, is_child = 0
1.01.741.401 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.01.741.401 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.01.741.408 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.01.741.409 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.01.741.414 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.01.741.414 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.01.741.422 I slot update_slots: id 3 | task 1447 | Checking checkpoint with [0, 108] against 0...
1.01.743.956 W slot update_slots: id 3 | task 1447 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.03.096.691 I slot print_timing: id 3 | task 1447 | prompt eval time = 245.49 ms / 22 tokens ( 11.16 ms per token, 89.62 tokens per second)
1.03.096.695 I slot print_timing: id 3 | task 1447 | eval time = 1109.77 ms / 70 tokens ( 15.85 ms per token, 63.08 tokens per second)
1.03.096.695 I slot print_timing: id 3 | task 1447 | total time = 1355.26 ms / 92 tokens
1.03.096.696 I slot print_timing: id 3 | task 1447 | graphs reused = 1444
1.03.096.730 I slot release: id 3 | task 1447 | stop processing: n_tokens = 199, truncated = 0
1.03.096.741 I srv update_slots: all slots are idle
1.03.104.983 I srv operator(): Chat format: peg-native
1.03.109.055 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.889 (> 0.100 thold), f_keep = 0.563
1.03.110.114 I slot launch_slot_: id 3 | task 1520 | processing task, is_child = 0
1.03.110.116 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.03.110.116 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.03.110.124 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.03.110.124 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.03.110.129 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.03.110.129 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.03.110.137 I slot update_slots: id 3 | task 1520 | Checking checkpoint with [0, 108] against 0...
1.03.112.735 W slot update_slots: id 3 | task 1520 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.04.099.696 I slot print_timing: id 3 | task 1520 | prompt eval time = 184.27 ms / 18 tokens ( 10.24 ms per token, 97.68 tokens per second)
1.04.099.699 I slot print_timing: id 3 | task 1520 | eval time = 805.28 ms / 51 tokens ( 15.79 ms per token, 63.33 tokens per second)
1.04.099.700 I slot print_timing: id 3 | task 1520 | total time = 989.55 ms / 69 tokens
1.04.099.701 I slot print_timing: id 3 | task 1520 | graphs reused = 1493
1.04.099.734 I slot release: id 3 | task 1520 | stop processing: n_tokens = 176, truncated = 0
1.04.099.742 I srv update_slots: all slots are idle
1.04.107.575 I srv operator(): Chat format: peg-native
1.04.111.251 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.862 (> 0.100 thold), f_keep = 0.636
1.04.112.105 I slot launch_slot_: id 3 | task 1574 | processing task, is_child = 0
1.04.112.106 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.04.112.107 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.04.112.114 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.04.112.114 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.04.112.119 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.04.112.119 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.04.112.127 I slot update_slots: id 3 | task 1574 | Checking checkpoint with [0, 108] against 0...
1.04.114.692 W slot update_slots: id 3 | task 1574 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.05.846.730 I slot print_timing: id 3 | task 1574 | prompt eval time = 241.09 ms / 22 tokens ( 10.96 ms per token, 91.25 tokens per second)
1.05.846.734 I slot print_timing: id 3 | task 1574 | eval time = 1493.51 ms / 94 tokens ( 15.89 ms per token, 62.94 tokens per second)
1.05.846.734 I slot print_timing: id 3 | task 1574 | total time = 1734.60 ms / 116 tokens
1.05.846.735 I slot print_timing: id 3 | task 1574 | graphs reused = 1585
1.05.846.768 I slot release: id 3 | task 1574 | stop processing: n_tokens = 223, truncated = 0
1.05.846.776 I srv update_slots: all slots are idle
1.05.854.609 I srv operator(): Chat format: peg-native
1.05.858.302 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.896 (> 0.100 thold), f_keep = 0.502
1.05.859.340 I slot launch_slot_: id 3 | task 1671 | processing task, is_child = 0
1.05.859.341 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.05.859.342 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.05.859.349 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.05.859.349 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.05.859.354 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.05.859.355 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.05.859.363 I slot update_slots: id 3 | task 1671 | Checking checkpoint with [0, 108] against 0...
1.05.861.934 W slot update_slots: id 3 | task 1671 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.07.056.737 I slot print_timing: id 3 | task 1671 | prompt eval time = 181.86 ms / 17 tokens ( 10.70 ms per token, 93.48 tokens per second)
1.07.056.741 I slot print_timing: id 3 | task 1671 | eval time = 1015.51 ms / 64 tokens ( 15.87 ms per token, 63.02 tokens per second)
1.07.056.741 I slot print_timing: id 3 | task 1671 | total time = 1197.37 ms / 81 tokens
1.07.056.742 I slot print_timing: id 3 | task 1671 | graphs reused = 1647
1.07.056.774 I slot release: id 3 | task 1671 | stop processing: n_tokens = 188, truncated = 0
1.07.056.781 I srv update_slots: all slots are idle
1.07.064.622 I srv operator(): Chat format: peg-native
1.07.068.316 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.903 (> 0.100 thold), f_keep = 0.596
1.07.069.348 I slot launch_slot_: id 3 | task 1738 | processing task, is_child = 0
1.07.069.350 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.07.069.351 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.07.069.358 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.07.069.359 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.07.069.364 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.07.069.364 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.07.069.372 I slot update_slots: id 3 | task 1738 | Checking checkpoint with [0, 108] against 0...
1.07.071.869 W slot update_slots: id 3 | task 1738 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.08.003.758 I slot print_timing: id 3 | task 1738 | prompt eval time = 176.17 ms / 16 tokens ( 11.01 ms per token, 90.82 tokens per second)
1.08.003.761 I slot print_timing: id 3 | task 1738 | eval time = 758.21 ms / 48 tokens ( 15.80 ms per token, 63.31 tokens per second)
1.08.003.762 I slot print_timing: id 3 | task 1738 | total time = 934.38 ms / 64 tokens
1.08.003.762 I slot print_timing: id 3 | task 1738 | graphs reused = 1693
1.08.003.792 I slot release: id 3 | task 1738 | stop processing: n_tokens = 171, truncated = 0
1.08.003.800 I srv update_slots: all slots are idle
1.08.011.689 I srv operator(): Chat format: peg-native
1.08.015.697 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.919 (> 0.100 thold), f_keep = 0.661
1.08.016.608 I slot launch_slot_: id 3 | task 1789 | processing task, is_child = 0
1.08.016.610 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.08.016.611 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.08.016.618 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.08.016.618 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.08.016.623 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.08.016.623 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.08.016.632 I slot update_slots: id 3 | task 1789 | Checking checkpoint with [0, 108] against 0...
1.08.019.281 W slot update_slots: id 3 | task 1789 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.09.008.769 I slot print_timing: id 3 | task 1789 | prompt eval time = 170.28 ms / 15 tokens ( 11.35 ms per token, 88.09 tokens per second)
1.09.008.772 I slot print_timing: id 3 | task 1789 | eval time = 821.85 ms / 52 tokens ( 15.80 ms per token, 63.27 tokens per second)
1.09.008.773 I slot print_timing: id 3 | task 1789 | total time = 992.13 ms / 67 tokens
1.09.008.774 I slot print_timing: id 3 | task 1789 | graphs reused = 1743
1.09.008.805 I slot release: id 3 | task 1789 | stop processing: n_tokens = 174, truncated = 0
1.09.008.814 I srv update_slots: all slots are idle
1.09.016.856 I srv operator(): Chat format: peg-native
1.09.020.607 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.918 (> 0.100 thold), f_keep = 0.644
1.09.021.583 I slot launch_slot_: id 3 | task 1844 | processing task, is_child = 0
1.09.021.585 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.09.021.586 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.09.021.593 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.09.021.593 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.09.021.598 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.09.021.599 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.09.021.606 I slot update_slots: id 3 | task 1844 | Checking checkpoint with [0, 108] against 0...
1.09.024.153 W slot update_slots: id 3 | task 1844 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.09.957.778 I slot print_timing: id 3 | task 1844 | prompt eval time = 159.46 ms / 14 tokens ( 11.39 ms per token, 87.80 tokens per second)
1.09.957.781 I slot print_timing: id 3 | task 1844 | eval time = 776.70 ms / 49 tokens ( 15.85 ms per token, 63.09 tokens per second)
1.09.957.782 I slot print_timing: id 3 | task 1844 | total time = 936.17 ms / 63 tokens
1.09.957.783 I slot print_timing: id 3 | task 1844 | graphs reused = 1790
1.09.957.813 I slot release: id 3 | task 1844 | stop processing: n_tokens = 170, truncated = 0
1.09.957.822 I srv update_slots: all slots are idle
1.09.965.471 I srv operator(): Chat format: peg-native
1.09.969.178 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.882 (> 0.100 thold), f_keep = 0.659
1.09.970.172 I slot launch_slot_: id 3 | task 1896 | processing task, is_child = 0
1.09.970.173 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.09.970.174 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.09.970.181 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.09.970.181 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.09.970.186 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.09.970.186 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.09.970.195 I slot update_slots: id 3 | task 1896 | Checking checkpoint with [0, 108] against 0...
1.09.972.830 W slot update_slots: id 3 | task 1896 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.11.296.814 I slot print_timing: id 3 | task 1896 | prompt eval time = 184.92 ms / 19 tokens ( 9.73 ms per token, 102.75 tokens per second)
1.11.296.818 I slot print_timing: id 3 | task 1896 | eval time = 1141.69 ms / 72 tokens ( 15.86 ms per token, 63.06 tokens per second)
1.11.296.818 I slot print_timing: id 3 | task 1896 | total time = 1326.61 ms / 91 tokens
1.11.296.819 I slot print_timing: id 3 | task 1896 | graphs reused = 1860
1.11.296.852 I slot release: id 3 | task 1896 | stop processing: n_tokens = 198, truncated = 0
1.11.296.861 I srv update_slots: all slots are idle
1.11.304.649 I srv operator(): Chat format: peg-native
1.11.308.613 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.889 (> 0.100 thold), f_keep = 0.566
1.11.309.502 I slot launch_slot_: id 3 | task 1971 | processing task, is_child = 0
1.11.309.504 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.11.309.505 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.11.309.511 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.11.309.511 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.11.309.516 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.11.309.517 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.11.309.525 I slot update_slots: id 3 | task 1971 | Checking checkpoint with [0, 108] against 0...
1.11.312.183 W slot update_slots: id 3 | task 1971 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.12.702.759 I slot print_timing: id 3 | task 1971 | prompt eval time = 187.47 ms / 18 tokens ( 10.41 ms per token, 96.02 tokens per second)
1.12.702.762 I slot print_timing: id 3 | task 1971 | eval time = 1205.76 ms / 76 tokens ( 15.87 ms per token, 63.03 tokens per second)
1.12.702.763 I slot print_timing: id 3 | task 1971 | total time = 1393.23 ms / 94 tokens
1.12.702.764 I slot print_timing: id 3 | task 1971 | graphs reused = 1934
1.12.702.794 I slot release: id 3 | task 1971 | stop processing: n_tokens = 201, truncated = 0
1.12.702.802 I srv update_slots: all slots are idle
1.12.710.667 I srv operator(): Chat format: peg-native
1.12.714.501 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.848 (> 0.100 thold), f_keep = 0.557
1.12.715.482 I slot launch_slot_: id 3 | task 2050 | processing task, is_child = 0
1.12.715.484 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.12.715.484 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.12.715.492 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.12.715.492 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.12.715.498 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.12.715.499 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.12.715.507 I slot update_slots: id 3 | task 2050 | Checking checkpoint with [0, 108] against 0...
1.12.718.107 W slot update_slots: id 3 | task 2050 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.14.519.810 I slot print_timing: id 3 | task 2050 | prompt eval time = 246.71 ms / 24 tokens ( 10.28 ms per token, 97.28 tokens per second)
1.14.519.813 I slot print_timing: id 3 | task 2050 | eval time = 1557.59 ms / 98 tokens ( 15.89 ms per token, 62.92 tokens per second)
1.14.519.814 I slot print_timing: id 3 | task 2050 | total time = 1804.30 ms / 122 tokens
1.14.519.815 I slot print_timing: id 3 | task 2050 | graphs reused = 2030
1.14.519.850 I slot release: id 3 | task 2050 | stop processing: n_tokens = 229, truncated = 0
1.14.519.858 I srv update_slots: all slots are idle
1.14.527.666 I srv operator(): Chat format: peg-native
1.14.531.540 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.855 (> 0.100 thold), f_keep = 0.489
1.14.531.542 I srv get_availabl: updating prompt cache
1.14.531.567 W srv prompt_save: - saving prompt with length 229, total state size = 12.556 MiB (draft: 0.000 MiB)
1.14.539.242 I srv load: - looking for better prompt, base f_keep = 0.489, sim = 0.855
1.14.539.252 I srv update: - cache state: 5 prompts, 87.187 MiB (limits: 8192.000 MiB, 8192 tokens, 121582 est)
1.14.539.252 I srv update: - prompt 0x65411f701010: 282 tokens, checkpoints: 1, 18.253 MiB
1.14.539.253 I srv update: - prompt 0x65411f8bbeb0: 253 tokens, checkpoints: 1, 17.233 MiB
1.14.539.253 I srv update: - prompt 0x65411c32e000: 285 tokens, checkpoints: 1, 18.359 MiB
1.14.539.254 I srv update: - prompt 0x65411f899dc0: 245 tokens, checkpoints: 1, 16.952 MiB
1.14.539.256 I srv update: - prompt 0x6541268d2b50: 229 tokens, checkpoints: 1, 16.389 MiB
1.14.539.258 I srv get_availabl: prompt cache update took 7.72 ms
1.14.540.191 I slot launch_slot_: id 3 | task 2151 | processing task, is_child = 0
1.14.540.193 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.14.540.194 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.14.540.203 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.14.540.204 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.14.540.209 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.14.540.209 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.14.540.222 I slot update_slots: id 3 | task 2151 | Checking checkpoint with [0, 108] against 0...
1.14.542.525 W slot update_slots: id 3 | task 2151 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.16.068.794 I slot print_timing: id 3 | task 2151 | prompt eval time = 241.67 ms / 23 tokens ( 10.51 ms per token, 95.17 tokens per second)
1.16.068.797 I slot print_timing: id 3 | task 2151 | eval time = 1286.89 ms / 81 tokens ( 15.89 ms per token, 62.94 tokens per second)
1.16.068.798 I slot print_timing: id 3 | task 2151 | total time = 1528.57 ms / 104 tokens
1.16.068.798 I slot print_timing: id 3 | task 2151 | graphs reused = 2109
1.16.068.830 I slot release: id 3 | task 2151 | stop processing: n_tokens = 211, truncated = 0
1.16.068.839 I srv update_slots: all slots are idle
1.16.076.793 I srv operator(): Chat format: peg-native
1.16.080.488 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.911 (> 0.100 thold), f_keep = 0.531
1.16.081.328 I slot launch_slot_: id 3 | task 2235 | processing task, is_child = 0
1.16.081.330 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.16.081.331 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.16.081.339 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.16.081.340 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.16.081.346 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.16.081.346 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.16.081.356 I slot update_slots: id 3 | task 2235 | Checking checkpoint with [0, 108] against 0...
1.16.083.922 W slot update_slots: id 3 | task 2235 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.16.993.808 I slot print_timing: id 3 | task 2235 | prompt eval time = 170.22 ms / 15 tokens ( 11.35 ms per token, 88.12 tokens per second)
1.16.993.813 I slot print_timing: id 3 | task 2235 | eval time = 742.23 ms / 47 tokens ( 15.79 ms per token, 63.32 tokens per second)
1.16.993.813 I slot print_timing: id 3 | task 2235 | total time = 912.45 ms / 62 tokens
1.16.993.814 I slot print_timing: id 3 | task 2235 | graphs reused = 2154
1.16.993.852 I slot release: id 3 | task 2235 | stop processing: n_tokens = 169, truncated = 0
1.16.993.864 I srv update_slots: all slots are idle
1.17.001.497 I srv operator(): Chat format: peg-native
1.17.005.259 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.889 (> 0.100 thold), f_keep = 0.663
1.17.006.136 I slot launch_slot_: id 3 | task 2285 | processing task, is_child = 0
1.17.006.139 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.17.006.139 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.17.006.147 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.17.006.147 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.17.006.152 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.17.006.152 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.17.006.160 I slot update_slots: id 3 | task 2285 | Checking checkpoint with [0, 108] against 0...
1.17.008.820 W slot update_slots: id 3 | task 2285 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.19.486.916 I slot print_timing: id 3 | task 2285 | prompt eval time = 181.65 ms / 18 tokens ( 10.09 ms per token, 99.09 tokens per second)
1.19.486.919 I slot print_timing: id 3 | task 2285 | eval time = 2299.10 ms / 144 tokens ( 15.97 ms per token, 62.63 tokens per second)
1.19.486.920 I slot print_timing: id 3 | task 2285 | total time = 2480.75 ms / 162 tokens
1.19.486.921 I slot print_timing: id 3 | task 2285 | graphs reused = 2295
1.19.486.952 I slot release: id 3 | task 2285 | stop processing: n_tokens = 269, truncated = 0
1.19.486.960 I srv update_slots: all slots are idle
1.19.494.973 I srv operator(): Chat format: peg-native
1.19.498.665 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.875 (> 0.100 thold), f_keep = 0.416
1.19.498.666 I srv get_availabl: updating prompt cache
1.19.498.691 W srv prompt_save: - saving prompt with length 269, total state size = 13.962 MiB (draft: 0.000 MiB)
1.19.512.453 I srv load: - looking for better prompt, base f_keep = 0.416, sim = 0.875
1.19.512.461 I srv update: - cache state: 6 prompts, 104.983 MiB (limits: 8192.000 MiB, 8192 tokens, 121963 est)
1.19.512.462 I srv update: - prompt 0x65411f701010: 282 tokens, checkpoints: 1, 18.253 MiB
1.19.512.462 I srv update: - prompt 0x65411f8bbeb0: 253 tokens, checkpoints: 1, 17.233 MiB
1.19.512.463 I srv update: - prompt 0x65411c32e000: 285 tokens, checkpoints: 1, 18.359 MiB
1.19.512.465 I srv update: - prompt 0x65411f899dc0: 245 tokens, checkpoints: 1, 16.952 MiB
1.19.512.466 I srv update: - prompt 0x6541268d2b50: 229 tokens, checkpoints: 1, 16.389 MiB
1.19.512.466 I srv update: - prompt 0x65411f927fd0: 269 tokens, checkpoints: 1, 17.796 MiB
1.19.512.468 I srv get_availabl: prompt cache update took 13.80 ms
1.19.513.421 I slot launch_slot_: id 3 | task 2432 | processing task, is_child = 0
1.19.513.423 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.19.513.424 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.19.513.431 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.19.513.431 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.19.513.436 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.19.513.437 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.19.513.448 I slot update_slots: id 3 | task 2432 | Checking checkpoint with [0, 108] against 0...
1.19.515.698 W slot update_slots: id 3 | task 2432 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.21.111.834 I slot print_timing: id 3 | task 2432 | prompt eval time = 201.14 ms / 20 tokens ( 10.06 ms per token, 99.43 tokens per second)
1.21.111.838 I slot print_timing: id 3 | task 2432 | eval time = 1397.24 ms / 88 tokens ( 15.88 ms per token, 62.98 tokens per second)
1.21.111.839 I slot print_timing: id 3 | task 2432 | total time = 1598.38 ms / 108 tokens
1.21.111.839 I slot print_timing: id 3 | task 2432 | graphs reused = 2381
1.21.111.872 I slot release: id 3 | task 2432 | stop processing: n_tokens = 215, truncated = 0
1.21.111.880 I srv update_slots: all slots are idle
1.21.119.478 I srv operator(): Chat format: peg-native
1.21.123.515 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.806 (> 0.100 thold), f_keep = 0.521
1.21.124.392 I slot launch_slot_: id 3 | task 2523 | processing task, is_child = 0
1.21.124.393 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.21.124.394 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.21.124.401 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.21.124.401 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.21.124.406 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.21.124.406 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.21.124.414 I slot update_slots: id 3 | task 2523 | Checking checkpoint with [0, 108] against 0...
1.21.127.004 W slot update_slots: id 3 | task 2523 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.24.453.140 I slot print_timing: id 3 | task 2523 | n_decoded = 188, tg = 62.66 t/s, tg_3s = 62.66 t/s
1.25.053.451 I slot print_timing: id 3 | task 2523 | prompt eval time = 328.40 ms / 31 tokens ( 10.59 ms per token, 94.40 tokens per second)
1.25.053.455 I slot print_timing: id 3 | task 2523 | eval time = 3600.63 ms / 226 tokens ( 15.93 ms per token, 62.77 tokens per second)
1.25.053.456 I slot print_timing: id 3 | task 2523 | total time = 3929.03 ms / 257 tokens
1.25.053.456 I slot print_timing: id 3 | task 2523 | graphs reused = 2604
1.25.053.492 I slot release: id 3 | task 2523 | stop processing: n_tokens = 364, truncated = 0
1.25.053.503 I srv update_slots: all slots are idle
1.25.061.456 I srv operator(): Chat format: peg-native
1.25.065.397 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.869 (> 0.100 thold), f_keep = 0.310
1.25.065.399 I srv get_availabl: updating prompt cache
1.25.065.438 W srv prompt_save: - saving prompt with length 364, total state size = 17.303 MiB (draft: 0.000 MiB)
1.25.081.705 I srv load: - looking for better prompt, base f_keep = 0.310, sim = 0.869
1.25.081.717 I srv update: - cache state: 7 prompts, 126.120 MiB (limits: 8192.000 MiB, 8192 tokens, 125165 est)
1.25.081.718 I srv update: - prompt 0x65411f701010: 282 tokens, checkpoints: 1, 18.253 MiB
1.25.081.719 I srv update: - prompt 0x65411f8bbeb0: 253 tokens, checkpoints: 1, 17.233 MiB
1.25.081.719 I srv update: - prompt 0x65411c32e000: 285 tokens, checkpoints: 1, 18.359 MiB
1.25.081.720 I srv update: - prompt 0x65411f899dc0: 245 tokens, checkpoints: 1, 16.952 MiB
1.25.081.723 I srv update: - prompt 0x6541268d2b50: 229 tokens, checkpoints: 1, 16.389 MiB
1.25.081.723 I srv update: - prompt 0x65411f927fd0: 269 tokens, checkpoints: 1, 17.796 MiB
1.25.081.724 I srv update: - prompt 0x65411f8618a0: 364 tokens, checkpoints: 1, 21.137 MiB
1.25.081.725 I srv get_availabl: prompt cache update took 16.32 ms
1.25.082.660 I slot launch_slot_: id 3 | task 2752 | processing task, is_child = 0
1.25.082.663 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.25.082.663 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.25.082.673 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.25.082.673 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.25.082.679 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.25.082.679 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.25.082.692 I slot update_slots: id 3 | task 2752 | Checking checkpoint with [0, 108] against 0...
1.25.085.058 W slot update_slots: id 3 | task 2752 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.27.580.446 I slot print_timing: id 3 | task 2752 | prompt eval time = 256.71 ms / 22 tokens ( 11.67 ms per token, 85.70 tokens per second)
1.27.580.449 I slot print_timing: id 3 | task 2752 | eval time = 2241.04 ms / 143 tokens ( 15.67 ms per token, 63.81 tokens per second)
1.27.580.450 I slot print_timing: id 3 | task 2752 | total time = 2497.75 ms / 165 tokens
1.27.580.451 I slot print_timing: id 3 | task 2752 | graphs reused = 2744
1.27.580.485 I slot release: id 3 | task 2752 | stop processing: n_tokens = 272, truncated = 0
1.27.580.493 I srv update_slots: all slots are idle
1.27.588.498 I srv operator(): Chat format: peg-native
1.27.592.239 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.896 (> 0.100 thold), f_keep = 0.412
1.27.592.241 I srv get_availabl: updating prompt cache
1.27.592.273 W srv prompt_save: - saving prompt with length 272, total state size = 14.068 MiB (draft: 0.000 MiB)
1.27.606.338 I srv load: - looking for better prompt, base f_keep = 0.412, sim = 0.896
1.27.606.349 I srv update: - cache state: 8 prompts, 144.022 MiB (limits: 8192.000 MiB, 8192 tokens, 125079 est)
1.27.606.350 I srv update: - prompt 0x65411f701010: 282 tokens, checkpoints: 1, 18.253 MiB
1.27.606.354 I srv update: - prompt 0x65411f8bbeb0: 253 tokens, checkpoints: 1, 17.233 MiB
1.27.606.354 I srv update: - prompt 0x65411c32e000: 285 tokens, checkpoints: 1, 18.359 MiB
1.27.606.355 I srv update: - prompt 0x65411f899dc0: 245 tokens, checkpoints: 1, 16.952 MiB
1.27.606.356 I srv update: - prompt 0x6541268d2b50: 229 tokens, checkpoints: 1, 16.389 MiB
1.27.606.356 I srv update: - prompt 0x65411f927fd0: 269 tokens, checkpoints: 1, 17.796 MiB
1.27.606.357 I srv update: - prompt 0x65411f8618a0: 364 tokens, checkpoints: 1, 21.137 MiB
1.27.606.360 I srv update: - prompt 0x65411f8c7630: 272 tokens, checkpoints: 1, 17.902 MiB
1.27.606.361 I srv get_availabl: prompt cache update took 14.12 ms
1.27.607.312 I slot launch_slot_: id 3 | task 2898 | processing task, is_child = 0
1.27.607.314 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.27.607.315 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.27.607.323 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.27.607.323 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.27.607.329 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.27.607.329 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.27.607.340 I slot update_slots: id 3 | task 2898 | Checking checkpoint with [0, 108] against 0...
1.27.609.682 W slot update_slots: id 3 | task 2898 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.30.315.227 I slot print_timing: id 3 | task 2898 | prompt eval time = 191.42 ms / 17 tokens ( 11.26 ms per token, 88.81 tokens per second)
1.30.315.230 I slot print_timing: id 3 | task 2898 | eval time = 2516.46 ms / 160 tokens ( 15.73 ms per token, 63.58 tokens per second)
1.30.315.231 I slot print_timing: id 3 | task 2898 | total time = 2707.88 ms / 177 tokens
1.30.315.231 I slot print_timing: id 3 | task 2898 | graphs reused = 2901
1.30.315.264 I slot release: id 3 | task 2898 | stop processing: n_tokens = 284, truncated = 0
1.30.315.273 I srv update_slots: all slots are idle
1.30.323.297 I srv operator(): Chat format: peg-native
1.30.327.225 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.862 (> 0.100 thold), f_keep = 0.394
1.30.327.227 I srv get_availabl: updating prompt cache
1.30.327.255 W srv prompt_save: - saving prompt with length 284, total state size = 14.490 MiB (draft: 0.000 MiB)
1.30.341.492 I srv load: - looking for better prompt, base f_keep = 0.394, sim = 0.862
1.30.341.499 I srv load: - found better prompt with f_keep = 0.401, sim = 0.869
1.30.347.490 I srv update: - cache state: 8 prompts, 144.092 MiB (limits: 8192.000 MiB, 8192 tokens, 125132 est)
1.30.347.492 I srv update: - prompt 0x65411f8bbeb0: 253 tokens, checkpoints: 1, 17.233 MiB
1.30.347.493 I srv update: - prompt 0x65411c32e000: 285 tokens, checkpoints: 1, 18.359 MiB
1.30.347.494 I srv update: - prompt 0x65411f899dc0: 245 tokens, checkpoints: 1, 16.952 MiB
1.30.347.494 I srv update: - prompt 0x6541268d2b50: 229 tokens, checkpoints: 1, 16.389 MiB
1.30.347.495 I srv update: - prompt 0x65411f927fd0: 269 tokens, checkpoints: 1, 17.796 MiB
1.30.347.495 I srv update: - prompt 0x65411f8618a0: 364 tokens, checkpoints: 1, 21.137 MiB
1.30.347.496 I srv update: - prompt 0x65411f8c7630: 272 tokens, checkpoints: 1, 17.902 MiB
1.30.347.496 I srv update: - prompt 0x65411f78f1e0: 284 tokens, checkpoints: 1, 18.324 MiB
1.30.347.498 I srv get_availabl: prompt cache update took 20.27 ms
1.30.348.408 I slot launch_slot_: id 3 | task 3061 | processing task, is_child = 0
1.30.348.410 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.30.348.411 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.30.348.420 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.30.348.420 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.30.348.425 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.30.348.425 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.30.348.438 I slot update_slots: id 3 | task 3061 | Checking checkpoint with [0, 108] against 0...
1.30.350.642 W slot update_slots: id 3 | task 3061 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.33.591.139 I slot print_timing: id 3 | task 3061 | n_decoded = 190, tg = 63.33 t/s, tg_3s = 63.33 t/s
1.33.956.468 I slot print_timing: id 3 | task 3061 | prompt eval time = 242.66 ms / 22 tokens ( 11.03 ms per token, 90.66 tokens per second)
1.33.956.471 I slot print_timing: id 3 | task 3061 | eval time = 3365.37 ms / 213 tokens ( 15.80 ms per token, 63.29 tokens per second)
1.33.956.471 I slot print_timing: id 3 | task 3061 | total time = 3608.03 ms / 235 tokens
1.33.956.472 I slot print_timing: id 3 | task 3061 | graphs reused = 3111
1.33.956.504 I slot release: id 3 | task 3061 | stop processing: n_tokens = 342, truncated = 0
1.33.956.512 I srv update_slots: all slots are idle
1.33.964.500 I srv operator(): Chat format: peg-native
1.33.968.210 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.850 (> 0.100 thold), f_keep = 0.330
1.33.968.212 I srv get_availabl: updating prompt cache
1.33.968.251 W srv prompt_save: - saving prompt with length 342, total state size = 16.530 MiB (draft: 0.000 MiB)
1.33.978.369 I srv load: - looking for better prompt, base f_keep = 0.330, sim = 0.850
1.33.978.378 I srv load: - found better prompt with f_keep = 0.404, sim = 0.865
1.33.984.495 I srv update: - cache state: 8 prompts, 146.097 MiB (limits: 8192.000 MiB, 8192 tokens, 126611 est)
1.33.984.500 I srv update: - prompt 0x65411f8bbeb0: 253 tokens, checkpoints: 1, 17.233 MiB
1.33.984.500 I srv update: - prompt 0x65411f899dc0: 245 tokens, checkpoints: 1, 16.952 MiB
1.33.984.501 I srv update: - prompt 0x6541268d2b50: 229 tokens, checkpoints: 1, 16.389 MiB
1.33.984.502 I srv update: - prompt 0x65411f927fd0: 269 tokens, checkpoints: 1, 17.796 MiB
1.33.984.502 I srv update: - prompt 0x65411f8618a0: 364 tokens, checkpoints: 1, 21.137 MiB
1.33.984.503 I srv update: - prompt 0x65411f8c7630: 272 tokens, checkpoints: 1, 17.902 MiB
1.33.984.504 I srv update: - prompt 0x65411f78f1e0: 284 tokens, checkpoints: 1, 18.324 MiB
1.33.984.505 I srv update: - prompt 0x65411f7a0fc0: 342 tokens, checkpoints: 1, 20.363 MiB
1.33.984.506 I srv get_availabl: prompt cache update took 16.29 ms
1.33.985.486 I slot launch_slot_: id 3 | task 3277 | processing task, is_child = 0
1.33.985.488 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.33.985.489 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.33.985.497 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.33.985.497 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.33.985.503 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.33.985.503 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.33.985.515 I slot update_slots: id 3 | task 3277 | Checking checkpoint with [0, 108] against 0...
1.33.987.748 W slot update_slots: id 3 | task 3277 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.35.834.417 I slot print_timing: id 3 | task 3277 | prompt eval time = 274.81 ms / 25 tokens ( 10.99 ms per token, 90.97 tokens per second)
1.35.834.421 I slot print_timing: id 3 | task 3277 | eval time = 1574.09 ms / 100 tokens ( 15.74 ms per token, 63.53 tokens per second)
1.35.834.421 I slot print_timing: id 3 | task 3277 | total time = 1848.90 ms / 125 tokens
1.35.834.422 I slot print_timing: id 3 | task 3277 | graphs reused = 3209
1.35.834.455 I slot release: id 3 | task 3277 | stop processing: n_tokens = 232, truncated = 0
1.35.834.464 I srv update_slots: all slots are idle
1.35.842.712 I srv operator(): Chat format: peg-native
1.35.846.428 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.896 (> 0.100 thold), f_keep = 0.483
1.35.846.430 I srv get_availabl: updating prompt cache
1.35.846.458 W srv prompt_save: - saving prompt with length 232, total state size = 12.661 MiB (draft: 0.000 MiB)
1.35.853.632 I srv load: - looking for better prompt, base f_keep = 0.483, sim = 0.896
1.35.853.644 I srv update: - cache state: 9 prompts, 162.592 MiB (limits: 8192.000 MiB, 8192 tokens, 125455 est)
1.35.853.645 I srv update: - prompt 0x65411f8bbeb0: 253 tokens, checkpoints: 1, 17.233 MiB
1.35.853.646 I srv update: - prompt 0x65411f899dc0: 245 tokens, checkpoints: 1, 16.952 MiB
1.35.853.650 I srv update: - prompt 0x6541268d2b50: 229 tokens, checkpoints: 1, 16.389 MiB
1.35.853.650 I srv update: - prompt 0x65411f927fd0: 269 tokens, checkpoints: 1, 17.796 MiB
1.35.853.651 I srv update: - prompt 0x65411f8618a0: 364 tokens, checkpoints: 1, 21.137 MiB
1.35.853.651 I srv update: - prompt 0x65411f8c7630: 272 tokens, checkpoints: 1, 17.902 MiB
1.35.853.652 I srv update: - prompt 0x65411f78f1e0: 284 tokens, checkpoints: 1, 18.324 MiB
1.35.853.652 I srv update: - prompt 0x65411f7a0fc0: 342 tokens, checkpoints: 1, 20.363 MiB
1.35.853.654 I srv update: - prompt 0x65411f788c50: 232 tokens, checkpoints: 1, 16.495 MiB
1.35.853.655 I srv get_availabl: prompt cache update took 7.22 ms
1.35.854.614 I slot launch_slot_: id 3 | task 3380 | processing task, is_child = 0
1.35.854.615 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.35.854.616 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.35.854.625 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.35.854.625 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.35.854.631 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.35.854.631 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.35.854.643 I slot update_slots: id 3 | task 3380 | Checking checkpoint with [0, 108] against 0...
1.35.856.942 W slot update_slots: id 3 | task 3380 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.37.310.429 I slot print_timing: id 3 | task 3380 | prompt eval time = 186.45 ms / 17 tokens ( 10.97 ms per token, 91.18 tokens per second)
1.37.310.432 I slot print_timing: id 3 | task 3380 | eval time = 1269.33 ms / 81 tokens ( 15.67 ms per token, 63.81 tokens per second)
1.37.310.433 I slot print_timing: id 3 | task 3380 | total time = 1455.78 ms / 98 tokens
1.37.310.433 I slot print_timing: id 3 | task 3380 | graphs reused = 3288
1.37.310.465 I slot release: id 3 | task 3380 | stop processing: n_tokens = 205, truncated = 0
1.37.310.474 I srv update_slots: all slots are idle
1.37.318.593 I srv operator(): Chat format: peg-native
1.37.322.354 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.926 (> 0.100 thold), f_keep = 0.546
1.37.323.237 I slot launch_slot_: id 3 | task 3464 | processing task, is_child = 0
1.37.323.239 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.37.323.240 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.37.323.246 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.37.323.246 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.37.323.252 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.37.323.252 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.37.323.260 I slot update_slots: id 3 | task 3464 | Checking checkpoint with [0, 108] against 0...
1.37.325.716 W slot update_slots: id 3 | task 3464 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.38.582.545 I slot print_timing: id 3 | task 3464 | prompt eval time = 89.00 ms / 13 tokens ( 6.85 ms per token, 146.08 tokens per second)
1.38.582.549 I slot print_timing: id 3 | task 3464 | eval time = 1170.29 ms / 74 tokens ( 15.81 ms per token, 63.23 tokens per second)
1.38.582.550 I slot print_timing: id 3 | task 3464 | total time = 1259.28 ms / 87 tokens
1.38.582.550 I slot print_timing: id 3 | task 3464 | graphs reused = 3360
1.38.582.584 I slot release: id 3 | task 3464 | stop processing: n_tokens = 194, truncated = 0
1.38.582.592 I srv update_slots: all slots are idle
1.38.590.412 I srv operator(): Chat format: peg-native
1.38.594.111 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.868 (> 0.100 thold), f_keep = 0.577
1.38.594.963 I slot launch_slot_: id 3 | task 3541 | processing task, is_child = 0
1.38.594.964 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.38.594.965 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.38.594.972 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.38.594.972 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.38.594.977 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.38.594.977 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.38.594.986 I slot update_slots: id 3 | task 3541 | Checking checkpoint with [0, 108] against 0...
1.38.597.490 W slot update_slots: id 3 | task 3541 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.40.441.102 I slot print_timing: id 3 | task 3541 | prompt eval time = 208.07 ms / 21 tokens ( 9.91 ms per token, 100.93 tokens per second)
1.40.441.105 I slot print_timing: id 3 | task 3541 | eval time = 1638.04 ms / 104 tokens ( 15.75 ms per token, 63.49 tokens per second)
1.40.441.106 I slot print_timing: id 3 | task 3541 | total time = 1846.11 ms / 125 tokens
1.40.441.106 I slot print_timing: id 3 | task 3541 | graphs reused = 3462
1.40.441.140 I slot release: id 3 | task 3541 | stop processing: n_tokens = 232, truncated = 0
1.40.441.149 I srv update_slots: all slots are idle
1.40.448.935 I srv operator(): Chat format: peg-native
1.40.452.707 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.896 (> 0.100 thold), f_keep = 0.483
1.40.452.709 I srv get_availabl: updating prompt cache
1.40.452.740 W srv prompt_save: - saving prompt with length 232, total state size = 12.661 MiB (draft: 0.000 MiB)
1.40.465.562 I srv load: - looking for better prompt, base f_keep = 0.483, sim = 0.896
1.40.465.568 I srv load: - found better prompt with f_keep = 0.496, sim = 0.920
1.40.471.040 I srv update: - cache state: 9 prompts, 162.592 MiB (limits: 8192.000 MiB, 8192 tokens, 125455 est)
1.40.471.042 I srv update: - prompt 0x65411f8bbeb0: 253 tokens, checkpoints: 1, 17.233 MiB
1.40.471.043 I srv update: - prompt 0x65411f899dc0: 245 tokens, checkpoints: 1, 16.952 MiB
1.40.471.043 I srv update: - prompt 0x6541268d2b50: 229 tokens, checkpoints: 1, 16.389 MiB
1.40.471.044 I srv update: - prompt 0x65411f927fd0: 269 tokens, checkpoints: 1, 17.796 MiB
1.40.471.044 I srv update: - prompt 0x65411f8618a0: 364 tokens, checkpoints: 1, 21.137 MiB
1.40.471.045 I srv update: - prompt 0x65411f8c7630: 272 tokens, checkpoints: 1, 17.902 MiB
1.40.471.045 I srv update: - prompt 0x65411f78f1e0: 284 tokens, checkpoints: 1, 18.324 MiB
1.40.471.046 I srv update: - prompt 0x65411f7a0fc0: 342 tokens, checkpoints: 1, 20.363 MiB
1.40.471.046 I srv update: - prompt 0x65411f786550: 232 tokens, checkpoints: 1, 16.495 MiB
1.40.471.047 I srv get_availabl: prompt cache update took 18.34 ms
1.40.471.974 I slot launch_slot_: id 3 | task 3648 | processing task, is_child = 0
1.40.471.976 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.40.471.977 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.40.471.986 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.40.471.986 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.40.471.991 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.40.471.992 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.40.472.005 I slot update_slots: id 3 | task 3648 | Checking checkpoint with [0, 108] against 0...
1.40.474.122 W slot update_slots: id 3 | task 3648 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.42.307.750 I slot print_timing: id 3 | task 3648 | prompt eval time = 186.55 ms / 17 tokens ( 10.97 ms per token, 91.13 tokens per second)
1.42.307.754 I slot print_timing: id 3 | task 3648 | eval time = 1649.19 ms / 105 tokens ( 15.71 ms per token, 63.67 tokens per second)
1.42.307.754 I slot print_timing: id 3 | task 3648 | total time = 1835.74 ms / 122 tokens
1.42.307.755 I slot print_timing: id 3 | task 3648 | graphs reused = 3565
1.42.307.791 I slot release: id 3 | task 3648 | stop processing: n_tokens = 229, truncated = 0
1.42.307.800 I srv update_slots: all slots are idle
1.42.315.625 I srv operator(): Chat format: peg-native
1.42.319.355 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.911 (> 0.100 thold), f_keep = 0.493
1.42.319.357 I srv get_availabl: updating prompt cache
1.42.319.387 W srv prompt_save: - saving prompt with length 229, total state size = 12.556 MiB (draft: 0.000 MiB)
1.42.326.475 I srv load: - looking for better prompt, base f_keep = 0.493, sim = 0.911
1.42.326.485 I srv update: - cache state: 10 prompts, 178.981 MiB (limits: 8192.000 MiB, 8192 tokens, 124448 est)
1.42.326.486 I srv update: - prompt 0x65411f8bbeb0: 253 tokens, checkpoints: 1, 17.233 MiB
1.42.326.487 I srv update: - prompt 0x65411f899dc0: 245 tokens, checkpoints: 1, 16.952 MiB
1.42.326.487 I srv update: - prompt 0x6541268d2b50: 229 tokens, checkpoints: 1, 16.389 MiB
1.42.326.488 I srv update: - prompt 0x65411f927fd0: 269 tokens, checkpoints: 1, 17.796 MiB
1.42.326.488 I srv update: - prompt 0x65411f8618a0: 364 tokens, checkpoints: 1, 21.137 MiB
1.42.326.489 I srv update: - prompt 0x65411f8c7630: 272 tokens, checkpoints: 1, 17.902 MiB
1.42.326.489 I srv update: - prompt 0x65411f78f1e0: 284 tokens, checkpoints: 1, 18.324 MiB
1.42.326.490 I srv update: - prompt 0x65411f7a0fc0: 342 tokens, checkpoints: 1, 20.363 MiB
1.42.326.490 I srv update: - prompt 0x65411f786550: 232 tokens, checkpoints: 1, 16.495 MiB
1.42.326.491 I srv update: - prompt 0x6541264cef50: 229 tokens, checkpoints: 1, 16.389 MiB
1.42.326.491 I srv get_availabl: prompt cache update took 7.13 ms
1.42.327.435 I slot launch_slot_: id 3 | task 3756 | processing task, is_child = 0
1.42.327.437 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.42.327.438 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.42.327.447 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.42.327.447 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.42.327.453 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.42.327.453 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.42.327.465 I slot update_slots: id 3 | task 3756 | Checking checkpoint with [0, 108] against 0...
1.42.329.839 W slot update_slots: id 3 | task 3756 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.44.626.425 I slot print_timing: id 3 | task 3756 | prompt eval time = 175.09 ms / 16 tokens ( 10.94 ms per token, 91.38 tokens per second)
1.44.626.428 I slot print_timing: id 3 | task 3756 | eval time = 2123.86 ms / 134 tokens ( 15.85 ms per token, 63.09 tokens per second)
1.44.626.429 I slot print_timing: id 3 | task 3756 | total time = 2298.95 ms / 150 tokens
1.44.626.429 I slot print_timing: id 3 | task 3756 | graphs reused = 3696
1.44.626.457 I slot release: id 3 | task 3756 | stop processing: n_tokens = 257, truncated = 0
1.44.626.467 I srv update_slots: all slots are idle
1.44.634.352 I srv operator(): Chat format: peg-native
1.44.638.063 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.882 (> 0.100 thold), f_keep = 0.436
1.44.638.065 I srv get_availabl: updating prompt cache
1.44.638.097 W srv prompt_save: - saving prompt with length 257, total state size = 13.540 MiB (draft: 0.000 MiB)
1.44.652.149 I srv load: - looking for better prompt, base f_keep = 0.436, sim = 0.882
1.44.652.158 I srv load: - found better prompt with f_keep = 0.493, sim = 0.890
1.44.657.896 I srv update: - cache state: 10 prompts, 179.966 MiB (limits: 8192.000 MiB, 8192 tokens, 125042 est)
1.44.657.899 I srv update: - prompt 0x65411f8bbeb0: 253 tokens, checkpoints: 1, 17.233 MiB
1.44.657.900 I srv update: - prompt 0x65411f899dc0: 245 tokens, checkpoints: 1, 16.952 MiB
1.44.657.900 I srv update: - prompt 0x65411f927fd0: 269 tokens, checkpoints: 1, 17.796 MiB
1.44.657.901 I srv update: - prompt 0x65411f8618a0: 364 tokens, checkpoints: 1, 21.137 MiB
1.44.657.901 I srv update: - prompt 0x65411f8c7630: 272 tokens, checkpoints: 1, 17.902 MiB
1.44.657.902 I srv update: - prompt 0x65411f78f1e0: 284 tokens, checkpoints: 1, 18.324 MiB
1.44.657.903 I srv update: - prompt 0x65411f7a0fc0: 342 tokens, checkpoints: 1, 20.363 MiB
1.44.657.903 I srv update: - prompt 0x65411f786550: 232 tokens, checkpoints: 1, 16.495 MiB
1.44.657.903 I srv update: - prompt 0x6541264cef50: 229 tokens, checkpoints: 1, 16.389 MiB
1.44.657.904 I srv update: - prompt 0x65411f909520: 257 tokens, checkpoints: 1, 17.374 MiB
1.44.657.905 I srv get_availabl: prompt cache update took 19.84 ms
1.44.658.807 I slot launch_slot_: id 3 | task 3893 | processing task, is_child = 0
1.44.658.808 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.44.658.809 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.44.658.817 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.44.658.817 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.44.658.822 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.44.658.822 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.44.658.833 I slot update_slots: id 3 | task 3893 | Checking checkpoint with [0, 108] against 0...
1.44.661.072 W slot update_slots: id 3 | task 3893 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.46.420.159 I slot print_timing: id 3 | task 3893 | prompt eval time = 187.75 ms / 19 tokens ( 9.88 ms per token, 101.20 tokens per second)
1.46.420.162 I slot print_timing: id 3 | task 3893 | eval time = 1573.57 ms / 99 tokens ( 15.89 ms per token, 62.91 tokens per second)
1.46.420.163 I slot print_timing: id 3 | task 3893 | total time = 1761.32 ms / 118 tokens
1.46.420.164 I slot print_timing: id 3 | task 3893 | graphs reused = 3793
1.46.420.201 I slot release: id 3 | task 3893 | stop processing: n_tokens = 225, truncated = 0
1.46.420.208 I srv update_slots: all slots are idle
1.46.427.888 I srv operator(): Chat format: peg-native
1.46.431.600 I slot get_availabl: id 3 | task -1 | selected slot by LCP similarity, sim_best = 0.911 (> 0.100 thold), f_keep = 0.498
1.46.431.602 I srv get_availabl: updating prompt cache
1.46.431.631 W srv prompt_save: - saving prompt with length 225, total state size = 12.415 MiB (draft: 0.000 MiB)
1.46.438.785 I srv load: - looking for better prompt, base f_keep = 0.498, sim = 0.911
1.46.438.792 I srv load: - found better prompt with f_keep = 0.498, sim = 0.927
1.46.444.375 I srv update: - cache state: 10 prompts, 179.825 MiB (limits: 8192.000 MiB, 8192 tokens, 124958 est)
1.46.444.377 I srv update: - prompt 0x65411f8bbeb0: 253 tokens, checkpoints: 1, 17.233 MiB
1.46.444.378 I srv update: - prompt 0x65411f899dc0: 245 tokens, checkpoints: 1, 16.952 MiB
1.46.444.378 I srv update: - prompt 0x65411f927fd0: 269 tokens, checkpoints: 1, 17.796 MiB
1.46.444.379 I srv update: - prompt 0x65411f8618a0: 364 tokens, checkpoints: 1, 21.137 MiB
1.46.444.380 I srv update: - prompt 0x65411f8c7630: 272 tokens, checkpoints: 1, 17.902 MiB
1.46.444.380 I srv update: - prompt 0x65411f78f1e0: 284 tokens, checkpoints: 1, 18.324 MiB
1.46.444.381 I srv update: - prompt 0x65411f7a0fc0: 342 tokens, checkpoints: 1, 20.363 MiB
1.46.444.381 I srv update: - prompt 0x65411f786550: 232 tokens, checkpoints: 1, 16.495 MiB
1.46.444.382 I srv update: - prompt 0x65411f909520: 257 tokens, checkpoints: 1, 17.374 MiB
1.46.444.382 I srv update: - prompt 0x65411f76f120: 225 tokens, checkpoints: 1, 16.249 MiB
1.46.444.383 I srv get_availabl: prompt cache update took 12.78 ms
1.46.445.328 I slot launch_slot_: id 3 | task 3995 | processing task, is_child = 0
1.46.445.330 I slot process_sing: id 0 | task -1 | saving idle slot to prompt cache
1.46.445.331 I slot prompt_clear: id 0 | task -1 | clearing prompt with 0 tokens
1.46.445.339 I slot process_sing: id 1 | task -1 | saving idle slot to prompt cache
1.46.445.339 I slot prompt_clear: id 1 | task -1 | clearing prompt with 0 tokens
1.46.445.344 I slot process_sing: id 2 | task -1 | saving idle slot to prompt cache
1.46.445.344 I slot prompt_clear: id 2 | task -1 | clearing prompt with 0 tokens
1.46.445.357 I slot update_slots: id 3 | task 3995 | Checking checkpoint with [0, 108] against 0...
1.46.447.602 W slot update_slots: id 3 | task 3995 | restored context checkpoint (pos_min = 0, pos_max = 108, n_tokens = 109, n_past = 108, size = 3.834 MiB)
1.48.345.464 I slot print_timing: id 3 | task 3995 | prompt eval time = 167.13 ms / 15 tokens ( 11.14 ms per token, 89.75 tokens per second)
1.48.345.467 I slot print_timing: id 3 | task 3995 | eval time = 1732.97 ms / 109 tokens ( 15.90 ms per token, 62.90 tokens per second)
1.48.345.468 I slot print_timing: id 3 | task 3995 | total time = 1900.10 ms / 124 tokens
1.48.345.469 I slot print_timing: id 3 | task 3995 | graphs reused = 3900
1.48.345.500 I slot release: id 3 | task 3995 | stop processing: n_tokens = 231, truncated = 0
1.48.345.509 I srv update_slots: all slots are idle
1.48.346.143 I srv operator(): operator(): cleaning up before exit...
Reference in New Issue Copy Permalink